How to Reproduce Our Work:

Step 1: Create GitHub Repo with necessary folders for partners to collaborate 00 Doc contains the .Rmd and HTML files 01 Data contains the data set that was worked with 02 Shiny contains the server and user interface (ui) for the Shiny App Step 2: Use an interesting data set with data that may be easily and interestingly manipulated Step 3: Import the csv file into SQL Developer Step 4: Save your app’s server.R and ui.R script inside the 02 Shiny folder Step 5: Launch the shinyApp with runApp and exit by clicking escape

Data Summary:

Below is the dataset and first ten rows

require("jsonlite")
## Loading required package: jsonlite
## 
## Attaching package: 'jsonlite'
## 
## The following object is masked from 'package:utils':
## 
##     View
require("RCurl")
## Loading required package: RCurl
## Loading required package: bitops
require(ggplot2)
## Loading required package: ggplot2
require(dplyr)
## Loading required package: dplyr
## 
## Attaching package: 'dplyr'
## 
## The following objects are masked from 'package:stats':
## 
##     filter, lag
## 
## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union
df <- data.frame(fromJSON(getURL(URLencode('skipper.cs.utexas.edu:5001/rest/native/?query="select * from edudata"'),httpheader=c(DB='jdbc:oracle:thin:@sayonara.microlab.cs.utexas.edu:1521:orcl', USER='C##cs329e_gmg954', PASS='orcl_gmg954', MODE='native_mode', MODEL='model', returnDimensions = 'False', returnFor = 'JSON'), verbose = TRUE)), ignoreNULL = FALSE)

head(df)
##   COUNTRY_NAME COUNTRY_CODE
## 1        Aruba          ABW
## 2        Aruba          ABW
## 3        Aruba          ABW
## 4        Aruba          ABW
## 5        Aruba          ABW
## 6        Aruba          ABW
##                                                      INDICATOR_NAME
## 1                       Lower secondary school starting age (years)
## 2 Lower secondary completion rate, female (% of relevant age group)
## 3   Lower secondary completion rate, male (% of relevant age group)
## 4  Lower secondary completion rate, total (% of relevant age group)
## 5                             Secondary education, duration (years)
## 6                                       Secondary education, pupils
##         INDICATOR_CODE            X2000            X2001            X2002
## 1          SE.SEC.AGES               12               12               12
## 2 SE.SEC.CMPT.LO.FE.ZS 96.4229965209961 100.603302001953  100.29239654541
## 3 SE.SEC.CMPT.LO.MA.ZS 93.1087036132812 95.9337005615234 94.1003036499023
## 4    SE.SEC.CMPT.LO.ZS 94.7530975341797 98.2667999267578 97.2099990844727
## 5          SE.SEC.DURS                5                5                5
## 6          SE.SEC.ENRL             6178             6428             6757
##              X2003            X2004            X2005 X2006
## 1               12               12               12    12
## 2 94.1010971069336 92.4527969360352 87.5162963867188  null
## 3 84.8787002563477 82.5789031982422 87.6658020019531  null
## 4 89.5258026123047 87.5595016479492 87.5903015136719  null
## 5                5                5                5     5
## 6             6869             6973             7116  7439
##              X2007 X2008            X2009 X2010 X2011 X2012 X2013 X2014
## 1               12    12               12    12    12    12    12    12
## 2 89.7855987548828  null 99.6067962646484  null  null  null  null  null
## 3 88.1463012695312  null 92.8752975463867  null  null  null  null  null
## 4 88.9660034179688  null 96.1911010742188  null  null  null  null  null
## 5                5     5                5     5     5     5     5     5
## 6             7853  7270             7439  7342  7378  8377  null  null
##   X2015 ignoreNULL
## 1  null      FALSE
## 2  null      FALSE
## 3  null      FALSE
## 4  null      FALSE
## 5  null      FALSE
## 6  null      FALSE

Data Explanation and Overal Manipulation:

Education is one of the most powerful instruments for reducing poverty and inequality and lays a foundation for sustained economic growth. The World Bank compiles data on education inputs, participation, efficiency, and outcomes. This data on education was compiled by the United Nations Educational, Scientific, and Cultural Organization (UNESCO) Institute for Statistics from official responses to surveys and from reports provided by education authorities in each country.

Below are all of the key indicators from our project:

require("jsonlite")
require("RCurl")
require(ggplot2)
require(dplyr)

df <- data.frame(fromJSON(getURL(URLencode('skipper.cs.utexas.edu:5001/rest/native/?query="select distinct(indicator_name) from edudata order by indicator_name desc"'),httpheader=c(DB='jdbc:oracle:thin:@sayonara.microlab.cs.utexas.edu:1521:orcl', USER='C##cs329e_gmg954', PASS='orcl_gmg954', MODE='native_mode', MODEL='model', returnDimensions = 'False', returnFor = 'JSON'), verbose = TRUE)), ignoreNULL = FALSE)

head(df)
##                                                          INDICATOR_NAME
## 1   Unemployment, total (% of total labor force) (modeled ILO estimate)
## 2     Unemployment, male (% of male labor force) (modeled ILO estimate)
## 3 Unemployment, female (% of female labor force) (modeled ILO estimate)
## 4      Trained teachers in primary education, male (% of male teachers)
## 5  Trained teachers in primary education, female (% of female teachers)
## 6           Trained teachers in primary education (% of total teachers)
##   ignoreNULL
## 1      FALSE
## 2      FALSE
## 3      FALSE
## 4      FALSE
## 5      FALSE
## 6      FALSE

Shiny Methodology

Shiny App Walkthrough:

Visualizations:

BoxPlot of Student-Teacher Ratios (Methodology Step One)

Data Summary:

This box plot shows the student-teacher ratios for the years 2000 and 2012 for both primary and secondary schools. We can see that in both years the student-teacher ratio for secondary schools is smaller than primary schools. We also can see that there has been an imporvement in student teacher ratios from 2000 to 2012. The interquartiles range has also decreased which means there is less spread in the data.

Process

  1. Place indicator name (student teacher ratio) on columns
  2. Place 2000 statistic and 2012 statistic on rows
  3. Add box plot from Show Me widget
  4. Color on 2000 statistic

Histogram of Student-Teacher Ratios (Methodology Step Two)

Data Summary:

This histogram shows the student teacher ratio’s in 2000. This graph shows that the student teacher ratios have a left skewed model. Most countries fall under the 10-30 ratio bin. However, some countries push extremely high ratios of 60 students per teacher. One country has a student teacher ratio of under 10.

Process

  1. Create a bin calculation with size 10
  2. Put (blue) bin on columns
  3. Place 2000 statistic on rows as count aggregate function
  4. Produces a histogram

Scatter Plot of Student Teacher Ratios (Methodology Step Three)

Data Summary:

(Also on Dashboard)

This scatterplot juxtaposes student teacher ratios from primary education to secondary education. From the years 2000 to 2012, we can see that values on the plot have a linear trend. The values that lie under the trend line depict an overall decrease in student teacher ratios (which is good). The average student teacher ratio for primary education is 27.7 and secondary educaiton is 18.2. This is expected because higher education usually tends to have more teachers per student.

Process

  1. 2002 statistic on column
  2. 2012 statistic on row
  3. Indicator name (unemployment rates) on columns
  4. Produces a scatter plot

SHINY SCATTER PLOT: https://graysong.shinyapps.io/Final_Project

CrossTab of Pupil-Teacher Ratios (Methodology Step Four)

Data Summary:

This crosstab shows which countries have some of the higher pupil-teacher ratio’s. Central African Republic has the highest ratio with about 80 students per teacher in primary education and 68 students per teacher in secondary education. This cross tab only shows those with ratios greater than 45 in primary education. The color represents the mins and maxes of the data displayed.

Process

  1. Indicator name on Column (Pupil-teacher ratio’s for primary and secondary education)
  2. Country name on rows
  3. Place agrregate 2012 statistic on text
  4. Filter country name by greater than a 40 student teacher ratio

Map of Student Teacher Ratios (Methodology Step Five)

Data Summary:

This map shows geographically where the lowest and highest pupil teacher ratio’s exist. We can see that the highest values come from African nations with other high values in Central America and South Eastern Asian countries. Europe, US, and East Asia have some of the lowest ratio’s, likely due to greatest economic development.

Process

  1. Place latitude and longitude on column and row
  2. Place 2012 statistic on to size and color
  3. Filter by indicator name student teacher ratio

Unemployment from 2002 to 2012

Data Summary:

This scatter plot shows the unemployment rates of several different countries. There are three plots which show male, female, and total unemployment rates of the countries. The X axis shows the unemployement rates for 2002 and the Y axis shows the unemployment rates for 2012. This means that the data above the line (Red), have increased unemployment rates and the data points below the line (Green) have decreased in unemployment rates. Female rates were more steady while male rate provided large variability.

Global Map of Number of Children Completing Primary School

Data Summary:

This map shows us the distribution of children who are not in primary school. The map shows that the vast majority of countries have very low rates of children out of school. There are however several somewhat surprising outliers. The United States has a much larger population of children out of school than any other North American or European country. While african countries do lag behind most other contients, Nigeria has a huge out-of-school child population, relativley speaking.

Time Progression of Percentage of Childreng Completing Primary School

Data Summary:

This graph shows a trend of countries that are classified by the United Nations as “Least Developed” in which more children are finishing their primary education more and more as the years go on. However, it appears that starting in 2011, the rate of children finishing their primary education seemes to be plateauing in the high 60% range. Before this shift, there are constant linear growth in this area. This may also shed some light on these countries and their corresponding wealth.

Number of children out of school from 2000 to 2012

SHINY BAR CHART: https://graysong.shinyapps.io/Final_Project

Cross-tab: Education Staff Compensation per Country

Data Summary:

This cross-tab shows the percentage of money alloted to education staff compenation that comes from total public institution expenditures in primary, secondary and tertiary public institutions. We can see that the largest percentages given to the education staff are in Niger, Guyana and Cameroon. This may be indicative of their overall economy. Niger, for example, has a relativley low spending percentage on primary education at 10%, but then spends a relativley huge amount on secondary education at 25%. Another interesting case is Guyana, which spends low amounts on primary and secondary education, but spends huge amounts on their tertiary education.

Male and Female Literacy Rates per Country

Data Summary:

This cross-tab shows the literacy rates amongst various countries’ male and female populations. Our results point to decreasing and increase percentages. For example, the Central Afrifan Republic has the largest decreasing literacy rate while Ghana has the largest increasing literacy rates. It can also be observed that the increases in literacy rates are larger in the female populations than in the male populations. This could be due to social change or more women entering school.

Process

  1. Indicator Name on Columns
  2. Country Name on rows
  3. Filter Indicator names
  4. Created Calculated Field –> Agg(Difference) from 2009 to 2010
  5. Filter at 3.5 percentage difference in all education levels

SHINY CROSS TAB: https://graysong.shinyapps.io/Final_Project

Government Expenditure Per Student

Data Summary:

This plot shows the average percentage of GDP per capita that is spent per student. These percentages are plotted as individual data points across the the years 2006 to 2012. It is intersting to see that government spending has only increased in tertiary education and remained constant in primary and secondary institutions. The importance of higher education has probably increased from 2006 to 2012.

Gross Intake Ratio for First Grade in Primary Education in Niger

Data Summary:

This visualization plots the gross intake ratio for first graders in Niger from the years 2000 to 2013. There is an increasing trend in intake and peaks in the year 2009. It is quite possible that in 2009 more money was allocated to Niger’s primary education, thus they were able to teach more primary students with an increased education budget.